Hacker News new | past | comments | ask | show | jobs | submit login
Google's New OS Will Offer Remote Desktop Capabilities (nytimes.com)
42 points by sound on June 13, 2010 | hide | past | favorite | 20 comments



Google is clearly making a play for the enterprise using small to medium businesses as a beachhead.

It has been noted here that the world's most widely used database is Microsoft Excel. This is because enterprise software is about top-down control of workflow and processes -- it's solidified corporate policy/politics! As a result, enterprise software sucks. In order to work around maladapted restrictions imposed by enterprise software, the most productive workers take their processes and automation into spreadsheets like Excel, where they can adapt quickly to changing conditions. Often, the most successful of these processes are then built into enterprise applications, to become the new status quo.

Google wants small and transitional medium sized businesses to outsource much of their IT to them and to run their spreadsheets on Google's apps. Google hopes that this will allow them to do with ad-hoc processes in spreadsheets what they did with web links: organize and unify a huge collection of disparate linkages to create value. When a company has all of its spreadsheets in Google apps, users will be able to link data in disparate spreadsheets. This network of links will be subject to analysis. New business processes will be absorbed from the ad-hoc spreadsheets into the enterprise more quickly and efficiently. This will have tremendous value to companies, especially small and medium sized ones, some of which will grow into giant conglomerates. If this works, Google stands to dwarf the achievements of Microsoft.

EDIT: This remote desktop stuff is just a selling point in support of larger goals.


I don't subscribe to your theory that the wide adoption of Excel as a database system is due to corporate politics. My experience (mostly in smaller companies of <1000 people) has shown that people use Excel because of low barriers (easy to learn), quick results, empowerment, and ownership. Corporate politics certainly plays a role in some environments, but if you eliminated politics, Excel would still dominate.


I don't subscribe to your theory that the wide adoption of Excel as a database system is due to corporate politics.

Well good, because that's not my theory. I never said Excel wasn't good on its own merits. What I am saying is that Excel is used for ad-hoc process and automation.

The phenomenon I actually described happens all the time on trading floors in the energy industry. Whatever trading system is in use gets in the way of something new. The processes to support the something new are run largely off of spreadsheets. If the something new is a big hit, higher-ups start a project to enable the new processes on the enterprise software where they can be better controlled.

My experience (mostly in smaller companies of <1000 people) has shown that people use Excel because of low barriers (easy to learn), quick results, empowerment, and ownership.

If Excel didn't have all of these attributes, it could never be used by non-programmers to support ad-hoc processes.


The trading example you mention illustrates how organizations can benefit by allowing people to innovate within the business units, see what works, and implement the successful ideas in a more scalable form.

One of the challenges is whether the person/people who developed the spreadsheet version and the enterprise development team can successfully manage a transition to a formal enterprise project. Sometimes, the people who build these spreadsheets see the job as their domain, and see outsiders as treading on their turf, while the development team is not always sensitive to the ownership issues.


The trading example you mention illustrates how organizations can benefit by allowing people to innovate within the business units, see what works, and implement the successful ideas in a more scalable form.

Most often, the organizations I've seen see this more as a problem of lack of control and wish they could stamp out spreadsheet use entirely.


After using them heavily for several years, I have sadly come to the conclusion that Google Docs and Spreadsheets are toys. Every time I try to get "serious" work done in them, I run into a show-stopping missing feature or performance bottleneck that leads me back to booting up a vm with Office installed.

Nobody in enterprise will take the Google office suite seriously unless there are some serious additions to performance and features.

Linking data from disparate spreadsheets sounds good in theory, until you find that trying to use any non-trivial dataset will break Google spreadsheets, your browser, or both.


After using them heavily for several years, I have sadly come to the conclusion that Google Docs and Spreadsheets are toys

I entirely agree. There's going to have to be a lot of heavy lifting to get them ready for what I described.


I think the most widely used database is sqlite.


If I install Ubuntu today (and probably other Linux distros) I get the option to do a RDP connection to apps or a desktop on a Windows server.

How is this different?


With your Ubuntu solution, does the computer you're logging into remotely have to be on? Do you have to know its IP address? I haven't used it before.

The advantage with this is to connect you probably simply visit chromeos.google.com (or something) and log-in. You don't have to worry about the IP address of the machine you're using or firewalls or anything like that.


Usually when I've seen people using remote desktop connections it is to access internal ERP or LOB applications - things which Google are not going to be able to provide access to.

For "normal" desktop apps (e.g. MS Office) - wouldn't providing access to these kinds of things undermine Google's strategy of doing things in the browser?

I guess they could provide a LogMeIn/GotoMyPC kind of product - but that is hardly revolutionary.


Yeah, I use the rdesktop command to remote in to the Windows machines at work. Works great!


Isn't Google just giving all these win32 apps a new lease on life? Forget HTML5. Just buy some Windows servers and move your win32 app to the cloud instead. It's the most practical thing to do if Google wants their OS running on lots of desktops but is it good for the future of the web? All I see here is more ad views for Google and more Windows licenses for Microsoft. We get stuck running win32 apps for the next decade because it's easier & cheaper than developing a native HTML5 app.


That's a fairly standard linux feature (X+SSH), isn't it? Seems like Chrome OS going to be the long awaited commercialization of Linux for the desktop.


A small note: X+SSH isn't very fast. Other software like VNC or NX generally work better, and Google wrote an open-source NX-compatible server:

http://code.google.com/p/neatx/


I think MacOS is close enough to a commercialization of Linux for the desktop to count as that. (I realize it's not based on Linux of course, but the result is very similar).


An OS that only runs on SSD's? commercialization of Linux for the desktop? I don't agree with you on this one.


To be honest, I am actually looking forward to install this OS into my system. I saw their feature video on this stating that turning Google Chrome bypasses the BIOS startup. A few of my colleagues said this is impossible, but for what it's worth, this is Google we're talking about.



it sounds somewhat like a vnc plugin in addition to their native chrome applications (if these 'legacy' apps indeed run 'in the cloud'), but could also include wine for windows apps or perhaps a nested X server for native linux apps. but the majority of the article is speculation and a review of recent tech history, so who knows




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: